
import os
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
sns.set(rc={'figure.figsize':(11, 4)})


path = 'C:/Users/Allard/Desktop/Projet_Call_SF' # Change to path to data
os.chdir(path)


df = pd.read_csv('Fire_Department_Calls_for_Service.csv')

C:\Users\Allard\anaconda3\envs\geopandas\lib\site-packages\IPython\core\interactiveshell.py:3146: DtypeWarning: Columns (19,20,25,30,34,39) have mixed types.Specify dtype option on import or set low_memory=False.
  has_raised = await self.run_ast_nodes(code_ast.body, cell_name,


df.columns

Index(['Call Number', 'Unit ID', 'Incident Number', 'Call Type', 'Call Date',
       'Watch Date', 'Received DtTm', 'Entry DtTm', 'Dispatch DtTm',
       'Response DtTm', 'On Scene DtTm', 'Transport DtTm', 'Hospital DtTm',
       'Call Final Disposition', 'Available DtTm', 'Address', 'City',
       'Zipcode of Incident', 'Battalion', 'Station Area', 'Box',
       'Original Priority', 'Priority', 'Final Priority', 'ALS Unit',
       'Call Type Group', 'Number of Alarms', 'Unit Type',
       'Unit sequence in call dispatch', 'Fire Prevention District',
       'Supervisor District', 'Neighborhooods - Analysis Boundaries',
       'Location', 'RowID', 'shape', 'Supervisor Districts',
       'Fire Prevention Districts', 'Current Police Districts',
       'Neighborhoods - Analysis Boundaries', 'Zip Codes',
       'Neighborhoods (old)', 'Police Districts',
       'Civic Center Harm Reduction Project Boundary', 'HSOC Zones',
       'Central Market/Tenderloin Boundary Polygon - Updated', 'Neighborhoods',
       'SF Find Neighborhoods', 'Current Police Districts 2',
       'Current Supervisor Districts'],
      dtype='object')


df.count()

Call Number                                             5459482
Unit ID                                                 5459482
Incident Number                                         5459482
Call Type                                               5459482
Call Date                                               5459482
Watch Date                                              5459482
Received DtTm                                           5459482
Entry DtTm                                              5459482
Dispatch DtTm                                           5459482
Response DtTm                                           5039441
On Scene DtTm                                           4233284
Transport DtTm                                          1440191
Hospital DtTm                                           1304525
Call Final Disposition                                  5459482
Available DtTm                                          5382474
Address                                                 5459482
City                                                    5451250
Zipcode of Incident                                     5445068
Battalion                                               5459482
Station Area                                            5457126
Box                                                     5459029
Original Priority                                       5433948
Priority                                                5459479
Final Priority                                          5459482
ALS Unit                                                5459482
Call Type Group                                         2647716
Number of Alarms                                        5459482
Unit Type                                               5459482
Unit sequence in call dispatch                          5459414
Fire Prevention District                                5459482
Supervisor District                                     5459482
Neighborhooods - Analysis Boundaries                    5459482
Location                                                5459482
RowID                                                   5459482
shape                                                   5361010
Supervisor Districts                                    5346843
Fire Prevention Districts                               5306039
Current Police Districts                                5326573
Neighborhoods - Analysis Boundaries                     5344884
Zip Codes                                               5360047
Neighborhoods (old)                                     5344884
Police Districts                                        5326573
Civic Center Harm Reduction Project Boundary             896614
HSOC Zones                                              1078356
Central Market/Tenderloin Boundary Polygon - Updated     934868
Neighborhoods                                           5315896
SF Find Neighborhoods                                   5315896
Current Police Districts 2                              5325114
Current Supervisor Districts                            5346843
dtype: int64


df1 = df.drop(columns=['Call Number', 'Unit ID', 'Incident Number',
       'Watch Date', 'Entry DtTm', 'Dispatch DtTm',
       'Response DtTm', 'On Scene DtTm', 'Transport DtTm', 'Hospital DtTm',
       'Call Final Disposition', 'Available DtTm', 'City',
       'Original Priority', 'Priority', 'ALS Unit',
       'Number of Alarms', 'Unit Type',
       'Unit sequence in call dispatch', 'Fire Prevention District',
       'Supervisor District', 'RowID', 'Supervisor Districts',
       'Fire Prevention Districts', 'Current Police Districts', 'Zip Codes',
       'Neighborhoods (old)', 'Police Districts',
       'Civic Center Harm Reduction Project Boundary', 'HSOC Zones',
       'Central Market/Tenderloin Boundary Polygon - Updated', 'Neighborhoods',
       'SF Find Neighborhoods', 'Current Police Districts 2',
       'Current Supervisor Districts'])


df1.count()

Call Type                               5459482
Call Date                               5459482
Received DtTm                           5459482
Address                                 5459482
Zipcode of Incident                     5445068
Battalion                               5459482
Station Area                            5457126
Box                                     5459029
Final Priority                          5459482
Call Type Group                         2647716
Neighborhooods - Analysis Boundaries    5459482
Location                                5459482
shape                                   5361010
Neighborhoods - Analysis Boundaries     5344884
dtype: int64


df1.columns

Index(['Call Type', 'Call Date', 'Received DtTm', 'Address',
       'Zipcode of Incident', 'Battalion', 'Station Area', 'Box',
       'Final Priority', 'Call Type Group',
       'Neighborhooods - Analysis Boundaries', 'Location', 'shape',
       'Neighborhoods - Analysis Boundaries'],
      dtype='object')


# ----- prend plusieurs minutes -----
#on transforme la colonne en type DateTime pour utiliser la colonne comme une date 
df1['Received DtTm'] = pd.to_datetime(df1['Received DtTm'], infer_datetime_format=True)
df1['Call Date'] = pd.to_datetime(df1['Call Date'], infer_datetime_format=True)


df2 = df1.copy()

df2['year'] = df2['Received DtTm'].dt.year
df2['month'] = df2['Received DtTm'].dt.month
df2['day'] = df2['Received DtTm'].dt.day
df2['hour'] = df2['Received DtTm'].dt.hour
df2['minute'] = df2['Received DtTm'].dt.minute
df2["weekday"] = df2['Received DtTm'].dt.day_name()


df2.columns

Index(['Call Type', 'Call Date', 'Received DtTm', 'Address',
       'Zipcode of Incident', 'Battalion', 'Station Area', 'Box',
       'Final Priority', 'Call Type Group',
       'Neighborhooods - Analysis Boundaries', 'Location', 'shape',
       'Neighborhoods - Analysis Boundaries', 'year', 'month', 'day', 'hour',
       'minute', 'weekday'],
      dtype='object')


df2.sort_values(by='Received DtTm', inplace=True)


df3 = df2.copy()


df3 = df3.drop_duplicates(subset=['Call Type', 'Call Date', 'Address',
       'Zipcode of Incident', 'Battalion', 'Final Priority', 'Call Type Group',
       'Neighborhooods - Analysis Boundaries', 'Location', 'year', 'month', 'day', 'hour', 'weekday'], keep='first')


df4 = df3.copy()


df4 = df4[(df4['Received DtTm'] > '2012-01-01') & (df4['Received DtTm'] < '2021-01-01')]


df4.loc[(df4["Call Type"] == "Medical Incident") & (df4["Call Type Group"] == "Non Life-threatening"), "Call Category"] = "Medical Non Life-Threatening"

df4.loc[(df4["Call Type"] == "Medical Incident") & (df4["Call Type Group"] == "Potentially Life-Threatening"), "Call Category"] = "Medical Potentially Life-Threatening"

df4.loc[df4["Call Type"] == "Outside Fire", "Call Category"] = "Fire"
df4.loc[df4["Call Type"] == "Structure Fire", "Call Category"] = "Fire"
df4.loc[df4["Call Type"] == "Vehicle Fire", "Call Category"] = "Fire"
df4.loc[df4["Call Type"] == "Marine Fire", "Call Category"] = "Fire"
df4.loc[df4["Call Type"] == "Train / Rail Fire", "Call Category"] = "Fire"
df4.loc[df4["Call Type"] == "Explosion", "Call Category"] = "Fire"

df4.loc[df4["Call Type"] == "Alarms", "Call Category"] = "Alarms"
df4.loc[(df4["Call Type"] == "Medical Incident") & (df4["Call Type Group"] == "Alarm"), "Call Category"] = "Alarms"

df4.loc[df4["Call Type"] == "Other", "Call Category"] = "Other"
df4.loc[df4["Call Type"] == "Administrative", "Call Category"] = "Other"
df4.loc[df4["Call Type"] == "Smoke Investigation (Outside)", "Call Category"] = "Other"
df4.loc[df4["Call Type"] == "Odor (Strange / Unknown)", "Call Category"] = "Other"
df4.loc[df4["Call Type"] == "Watercraft in Distress", "Call Category"] = "Other"
df4.loc[df4["Call Type"] == "Confined Space / Structure Collapse", "Call Category"] = "Other"
df4.loc[df4["Call Type"] == "Extrication / Entrapped (Machinery, Vehicle)", "Call Category"] = "Other"
df4.loc[df4["Call Type"] == "Aircraft Emergency", "Call Category"] = "Other"
df4.loc[df4["Call Type"] == "Suspicious Package", "Call Category"] = "Other"
df4.loc[df4["Call Type"] == "Citizen Assist / Service Call", "Call Category"] = "Other"
df4.loc[df4["Call Type"] == "Assist Police", "Call Category"] = "Other"
df4.loc[df4["Call Type"] == "Mutual Aid / Assist Outside Agency", "Call Category"] = "Other"

df4.loc[df4["Call Type"] == "Electrical Hazard", "Call Category"] = "Incidents"
df4.loc[df4["Call Type"] == "Gas Leak (Natural and LP Gases)", "Call Category"] = "Incidents"
df4.loc[df4["Call Type"] == "Fuel Spill", "Call Category"] = "Incidents"
df4.loc[df4["Call Type"] == "HazMat", "Call Category"] = "Incidents"
df4.loc[df4["Call Type"] == "Oil Spill", "Call Category"] = "Incidents"
df4.loc[df4["Call Type"] == "Industrial Accidents", "Call Category"] = "Incidents"
df4.loc[df4["Call Type"] == "Train / Rail Incident", "Call Category"] = "Incidents"
df4.loc[df4["Call Type"] == "Traffic Collision", "Call Category"] = "Incidents"
df4.loc[df4["Call Type"] == "High Angle Rescue", "Call Category"] = "Incidents"
df4.loc[df4["Call Type"] == "Water Rescue", "Call Category"] = "Incidents"
df4.loc[df4["Call Type"] == "Elevator / Escalator Rescue", "Call Category"] = "Incidents"


df4["Call Category"].unique()

array(['Medical Non Life-Threatening', 'Fire',
       'Medical Potentially Life-Threatening', 'Incidents', 'Other',
       'Alarms', nan], dtype=object)


df4.drop(columns=["Call Type","Call Type Group"], inplace=True)


df4.count()

Call Date                               1227726
Received DtTm                           1227726
Address                                 1227726
Zipcode of Incident                     1224663
Battalion                               1227726
Station Area                            1226615
Box                                     1227649
Final Priority                          1227726
Neighborhooods - Analysis Boundaries    1227726
Location                                1227726
shape                                   1182755
Neighborhoods - Analysis Boundaries     1179441
year                                    1227726
month                                   1227726
day                                     1227726
hour                                    1227726
minute                                  1227726
weekday                                 1227726
Call Category                           1223520
dtype: int64


df4.dropna(inplace=True)


df4.count()

Call Date                               1175862
Received DtTm                           1175862
Address                                 1175862
Zipcode of Incident                     1175862
Battalion                               1175862
Station Area                            1175862
Box                                     1175862
Final Priority                          1175862
Neighborhooods - Analysis Boundaries    1175862
Location                                1175862
shape                                   1175862
Neighborhoods - Analysis Boundaries     1175862
year                                    1175862
month                                   1175862
day                                     1175862
hour                                    1175862
minute                                  1175862
weekday                                 1175862
Call Category                           1175862
dtype: int64


df4.head()


# Importation de notre nouvelle base de données que nous utiliserons dans les parties 2 et 3
df4.to_csv('Fire_Department_Calls_for_Service_reduced_V3.csv')


df4["Call Category"].unique()

array(['Medical Non Life-Threatening', 'Fire',
       'Medical Potentially Life-Threatening', 'Incidents', 'Other',
       'Alarms'], dtype=object)


plt.figure(figsize=(15,5))
sns.countplot(x='Call Category', data=df4, 
              order = df4['Call Category'].value_counts().index)
plt.xticks(rotation=90)
plt.title("Nombre d'appels par catégorie")
plt.show()


def donnees2(nom):
    return df4.loc[df4["Call Category"] == nom]


dataMNLT = donnees2("Medical Non Life-Threatening")
dataMNLT.head(5)


r = pd.DataFrame(dataMNLT["month"].value_counts())
r


r["Mois"] = r.index
r.columns = ["Effectif", "Mois"]
r


sns.lineplot(data = r, x = "Mois", y = "Effectif").set_title('Medical Non Life-Threatening')

Text(0.5, 1.0, 'Medical Non Life-Threatening')


dataMPLT = donnees2("Medical Potentially Life-Threatening")
rh = pd.DataFrame(dataMPLT["month"].value_counts())
rh["Mois"] = rh.index
rh.columns = ["Effectif", "Mois"]
sns.lineplot(data = rh, x = "Mois", y = "Effectif").set_title('Medical Potentially Life-Threatening')

Text(0.5, 1.0, 'Medical Potentially Life-Threatening')


fig = plt.figure(figsize=(20,25))
for i,nom in enumerate(df4["Call Category"].unique(), 0):
    datamed = donnees2(nom)
    raa = pd.DataFrame(datamed["year"].value_counts())
    raa["Année"] = raa.index
    raa.columns = ["Effectif", "Année"]
    ax = fig.add_subplot(4, 3, i+1)
    sns.lineplot(data = raa, x = "Année", y = "Effectif", ax = ax)    
    ax.set_title(nom)


fig = plt.figure(figsize=(20,25))
for i,nom in enumerate(df4["Call Category"].unique(), 0):
    datamed = donnees2(nom)
    rr = pd.DataFrame(datamed["month"].value_counts())
    rr["Mois"] = rr.index
    rr.columns = ["Effectif", "Mois"]
    ax = fig.add_subplot(4, 3, i+1)
    sns.lineplot(data = rr, x = "Mois", y = "Effectif", ax = ax)    
    ax.set_title(nom)


fig = plt.figure(figsize=(20,25))
for i,nom in enumerate(df4["Call Category"].unique(), 0):
    datamed = donnees2(nom)
    rr = pd.DataFrame(datamed["day"].value_counts())
    rr["Jour du mois"] = rr.index
    rr.columns = ["Effectif", "Jour du mois"]
    ax = fig.add_subplot(4, 3, i+1)
    sns.lineplot(data = rr, x = "Jour du mois", y = "Effectif", ax = ax)    
    ax.set_title(nom)


fig = plt.figure(figsize=(25,15))
for i,nom in enumerate(df4["Call Category"].unique(), 0):  
    data_req = df4[(df4['Call Category'] == nom)].groupby(['day', 'weekday']).size().unstack(level=1)
    ax = fig.add_subplot(2, 3, i+1)
    data_req.plot(kind='bar', stacked=True, ax=ax)
    plt.title(nom)
    plt.xlabel('day', fontsize=10)
    plt.xticks(rotation=1)
    #plt.legend(bbox_to_anchor=(1.01, 1),borderaxespad=0)
    plt.ylabel('Count', fontsize=10)
    ax.get_legend().remove()
plt.legend(bbox_to_anchor=(1.01, 1),borderaxespad=0)

<matplotlib.legend.Legend at 0x206e0d5de80>


fig = plt.figure(figsize=(25,15))
for i,nom in enumerate(df4["Call Category"].unique(), 0):  
    data_req = df4[(df4['Call Category'] == nom)].groupby(['weekday', 'year']).size().unstack(level=1)
    ax = fig.add_subplot(2, 3, i+1)
    data_req.plot(kind='bar', stacked=True, ax=ax)
    plt.title(nom)
    plt.xlabel('day', fontsize=10)
    plt.xticks(rotation=1)
    #plt.legend(bbox_to_anchor=(1.01, 1),borderaxespad=0)
    plt.ylabel('Count', fontsize=10)
    ax.get_legend().remove()
plt.legend(bbox_to_anchor=(1.01, 1),borderaxespad=0)

<matplotlib.legend.Legend at 0x206209f14c0>


fig = plt.figure(figsize=(20,25))
for i,nom in enumerate(df4["Call Category"].unique(), 0):
    datamed = donnees2(nom)
    ra = pd.DataFrame(datamed["hour"].value_counts())
    ra["Heure"] = ra.index
    ra.columns = ["Effectif", "Heure"]
    ax = fig.add_subplot(4, 3, i+1)
    sns.lineplot(data = ra, x = "Heure", y = "Effectif", ax = ax, legend="full")    
    ax.set_title(nom)


fig = plt.figure(figsize=(25,15))
for i,nom in enumerate(df4["Call Category"].unique(), 0):  
    data_req = df4[(df4['Call Category'] == nom)].groupby(['hour', 'weekday']).size().unstack(level=1)
    ax = fig.add_subplot(2, 3, i+1)
    data_req.plot(kind='bar', stacked=True, ax=ax)
    plt.title(nom)
    plt.xlabel('hour', fontsize=10)
    plt.xticks(rotation=1)
    #plt.legend(bbox_to_anchor=(1.01, 1),borderaxespad=0)
    plt.ylabel('Count', fontsize=10)
    #ax.get_legend().remove()
#plt.legend(bbox_to_anchor=(1.01, 1),borderaxespad=0)


palette ={"Medical Potentially Life-Threatening": "crimson", "Medical Non Life-Threatening": "dodgerblue", "Fire": "mediumseagreen", "Alarms": "darkorange", "Other": "aqua", "Incidents": "dimgrey"}

fig = plt.figure(figsize=(20,15))
for i in range(0,9):
    ax = fig.add_subplot(3, 3, i+1)
    sns.countplot(x='month', hue='Call Category', palette=palette, data=df4.loc[df4['year']==2012+i,:], ax=ax)
    ax.set_title('Calls per month '+str(2012+i))
    ax.get_legend().remove()
plt.legend(bbox_to_anchor=(1.01, 1),borderaxespad=0)

<matplotlib.legend.Legend at 0x2060bbe3700>


df4["Neighborhooods - Analysis Boundaries"].unique().shape

(41,)


# Pourcentage des incidents par quartier
df4["Neighborhooods - Analysis Boundaries"].value_counts(normalize=True) * 100

Tenderloin                        14.442511
South of Market                   10.942185
Mission                            9.502731
Financial District/South Beach     6.857692
Bayview Hunters Point              5.140994
Sunset/Parkside                    3.724247
Western Addition                   3.537490
Nob Hill                           3.156153
Castro/Upper Market                2.657200
Hayes Valley                       2.559144
Outer Richmond                     2.409211
West of Twin Peaks                 2.119892
North Beach                        2.032296
Excelsior                          1.996153
Chinatown                          1.929648
Pacific Heights                    1.862889
Marina                             1.819091
Potrero Hill                       1.716868
Bernal Heights                     1.670519
Haight Ashbury                     1.588707
Inner Sunset                       1.449575
Mission Bay                        1.443112
Lakeshore                          1.430015
Outer Mission                      1.429760
Russian Hill                       1.385622
Oceanview/Merced/Ingleside         1.280848
Inner Richmond                     1.201672
Visitacion Valley                  1.171566
Lone Mountain/USF                  1.152771
Noe Valley                         0.985915
Portola                            0.954024
Japantown                          0.891261
Presidio Heights                   0.763440
Golden Gate Park                   0.596328
Twin Peaks                         0.580680
Treasure Island                    0.487642
Glen Park                          0.466551
Presidio                           0.359906
Seacliff                           0.132839
McLaren Park                       0.102393
Lincoln Park                       0.068460
Name: Neighborhooods - Analysis Boundaries, dtype: float64


plt.figure(figsize=(15,5))
sns.countplot(x='Neighborhooods - Analysis Boundaries', data=df4, 
              order = df4['Neighborhooods - Analysis Boundaries'].value_counts().index)
plt.xticks(rotation=90)
plt.title("Nombre d'appels par quartier")
plt.show()


data_req = df4[(df4['Neighborhooods - Analysis Boundaries'] == 'Tenderloin')].groupby(['Call Category', 'weekday']).size().unstack(level=1)
data_req.plot(kind='bar', stacked=True, figsize=(6,6))
plt.title("Nombre d'appels par catégorie à Tenderloin")
plt.xlabel('Call Category', fontsize=10)
plt.xticks(rotation=90)
plt.legend(bbox_to_anchor=(1.01, 1),borderaxespad=0)
plt.ylabel('Count', fontsize=10)

Text(0, 0.5, 'Count')


data_req = df4[(df4['Neighborhooods - Analysis Boundaries'] == 'Lincoln Park')].groupby(['Call Category', 'weekday']).size().unstack(level=1)
data_req.plot(kind='bar', stacked=True, figsize=(6,6))
plt.title("Nombre d'appels par catégorie à Lincoln Park")
plt.xlabel('Call Category', fontsize=10)
plt.xticks(rotation=90)
plt.legend(bbox_to_anchor=(1.01, 1),borderaxespad=0)
plt.ylabel('Count', fontsize=10)

Text(0, 0.5, 'Count')


!pip install folium
import folium

Requirement already satisfied: folium in c:\users\allard\anaconda3\envs\geopandas\lib\site-packages (0.12.1)
Requirement already satisfied: numpy in c:\users\allard\anaconda3\envs\geopandas\lib\site-packages (from folium) (1.19.2)
Requirement already satisfied: branca>=0.3.0 in c:\users\allard\anaconda3\envs\geopandas\lib\site-packages (from folium) (0.4.2)
Requirement already satisfied: jinja2>=2.9 in c:\users\allard\anaconda3\envs\geopandas\lib\site-packages (from folium) (2.11.2)
Requirement already satisfied: requests in c:\users\allard\anaconda3\envs\geopandas\lib\site-packages (from folium) (2.25.1)
Requirement already satisfied: MarkupSafe>=0.23 in c:\users\allard\anaconda3\envs\geopandas\lib\site-packages (from jinja2>=2.9->folium) (1.1.1)
Requirement already satisfied: urllib3<1.27,>=1.21.1 in c:\users\allard\anaconda3\envs\geopandas\lib\site-packages (from requests->folium) (1.26.3)
Requirement already satisfied: certifi>=2017.4.17 in c:\users\allard\anaconda3\envs\geopandas\lib\site-packages (from requests->folium) (2020.12.5)
Requirement already satisfied: chardet<5,>=3.0.2 in c:\users\allard\anaconda3\envs\geopandas\lib\site-packages (from requests->folium) (4.0.0)
Requirement already satisfied: idna<3,>=2.5 in c:\users\allard\anaconda3\envs\geopandas\lib\site-packages (from requests->folium) (2.10)


sf = df4.groupby('Neighborhooods - Analysis Boundaries').count()
sf = pd.DataFrame(sf,columns=['Call Category'])  # remove unneeded columns
sf.reset_index(inplace=True)   # default index, otherwise groupby column becomes index
sf.rename(columns={'Neighborhooods - Analysis Boundaries':'Neighborhood','Call Category':'Count'}, inplace=True)
sf.sort_values(by='Count', inplace=True, ascending=False)


# San Francisco latitude and longitude values
latitude = 37.77
longitude = -122.42
sf_neighborhood_geo = 'Analysis Neighborhoods.geojson'

# Create map
sf_map = folium.Map(
       location=[latitude,longitude],
       zoom_start=12)

 
# Use json file  TEST based on class
sf_map.choropleth(
       geo_data=sf_neighborhood_geo,
       data=sf,
       columns=['Neighborhood','Count'],
       key_on='feature.properties.nhood',
       fill_color='YlOrRd',
       fill_opacity='0.7',
       line_opacity='0.2',
       legend_name='Appels à San Francisco, par quartier')

# display the map
sf_map

C:\Users\Allard\anaconda3\envs\geopandas\lib\site-packages\folium\folium.py:409: FutureWarning: The choropleth  method has been deprecated. Instead use the new Choropleth class, which has the same arguments. See the example notebook 'GeoJSON_and_choropleth' for how to do this.
  warnings.warn(


df4_fire=df4.loc[df4['Call Category']=='Fire']
sf = df4_fire.groupby('Neighborhooods - Analysis Boundaries').count()
sf = pd.DataFrame(sf,columns=['Call Category'])  # remove unneeded columns
sf.reset_index(inplace=True)   # default index, otherwise groupby column becomes index
sf.rename(columns={'Neighborhooods - Analysis Boundaries':'Neighborhood','Call Category':'Count'}, inplace=True)
sf.sort_values(by='Count', inplace=True, ascending=False)


# San Francisco latitude and longitude values
latitude = 37.77
longitude = -122.42
sf_neighborhood_geo = 'Analysis Neighborhoods.geojson'

# Create map
sf_map = folium.Map(
       location=[latitude,longitude],
       zoom_start=12)

 
# Use json file  TEST based on class
sf_map.choropleth(
       geo_data=sf_neighborhood_geo,
       data=sf,
       columns=['Neighborhood','Count'],
       key_on='feature.properties.nhood',
       fill_color='YlOrRd',
       fill_opacity='0.7',
       line_opacity='0.2',
       legend_name='Urgences Fire à San Francisco, par quartier')

# display the map
sf_map


df4['Call Category'].unique()

array(['Medical Non Life-Threatening', 'Fire',
       'Medical Potentially Life-Threatening', 'Incidents', 'Other',
       'Alarms'], dtype=object)


df4_fire=df4.loc[df4['Call Category']=='Other']
sf = df4_fire.groupby('Neighborhooods - Analysis Boundaries').count()
sf = pd.DataFrame(sf,columns=['Call Category'])  # remove unneeded columns
sf.reset_index(inplace=True)   # default index, otherwise groupby column becomes index
sf.rename(columns={'Neighborhooods - Analysis Boundaries':'Neighborhood','Call Category':'Count'}, inplace=True)
sf.sort_values(by='Count', inplace=True, ascending=False)


# San Francisco latitude and longitude values
latitude = 37.77
longitude = -122.42
sf_neighborhood_geo = 'Analysis Neighborhoods.geojson'

# Create map
sf_map = folium.Map(
       location=[latitude,longitude],
       zoom_start=12)

 
# Use json file  TEST based on class
sf_map.choropleth(
       geo_data=sf_neighborhood_geo,
       data=sf,
       columns=['Neighborhood','Count'],
       key_on='feature.properties.nhood',
       fill_color='YlOrRd',
       fill_opacity='0.7',
       line_opacity='0.2',
       legend_name='Urgences Other à San Francisco, par quartier')

# display the map
sf_map


df4['Call Category'].unique()

array(['Medical Non Life-Threatening', 'Fire',
       'Medical Potentially Life-Threatening', 'Incidents', 'Other',
       'Alarms'], dtype=object)


df4_fire=df4.loc[df4['Call Category']=='Medical Non Life-Threatening']
sf = df4_fire.groupby('Neighborhooods - Analysis Boundaries').count()
sf = pd.DataFrame(sf,columns=['Call Category'])  # remove unneeded columns
sf.reset_index(inplace=True)   # default index, otherwise groupby column becomes index
sf.rename(columns={'Neighborhooods - Analysis Boundaries':'Neighborhood','Call Category':'Count'}, inplace=True)
sf.sort_values(by='Count', inplace=True, ascending=False)


# San Francisco latitude and longitude values
latitude = 37.77
longitude = -122.42
sf_neighborhood_geo = 'Analysis Neighborhoods.geojson'

# Create map
sf_map = folium.Map(
       location=[latitude,longitude],
       zoom_start=12)

 
# Use json file  TEST based on class
sf_map.choropleth(
       geo_data=sf_neighborhood_geo,
       data=sf,
       columns=['Neighborhood','Count'],
       key_on='feature.properties.nhood',
       fill_color='YlOrRd',
       fill_opacity='0.7',
       line_opacity='0.2',
       legend_name='Urgences Medical Non Life-Threatening à San Francisco, par quartier')

# display the map
sf_map


df4_fire=df4.loc[df4['Call Category']=='Medical Potentially Life-Threatening']
sf = df4_fire.groupby('Neighborhooods - Analysis Boundaries').count()
sf = pd.DataFrame(sf,columns=['Call Category'])  # remove unneeded columns
sf.reset_index(inplace=True)   # default index, otherwise groupby column becomes index
sf.rename(columns={'Neighborhooods - Analysis Boundaries':'Neighborhood','Call Category':'Count'}, inplace=True)
sf.sort_values(by='Count', inplace=True, ascending=False)


# San Francisco latitude and longitude values
latitude = 37.77
longitude = -122.42
sf_neighborhood_geo = 'Analysis Neighborhoods.geojson'

# Create map
sf_map = folium.Map(
       location=[latitude,longitude],
       zoom_start=12)

 
# Use json file  TEST based on class
sf_map.choropleth(
       geo_data=sf_neighborhood_geo,
       data=sf,
       columns=['Neighborhood','Count'],
       key_on='feature.properties.nhood',
       fill_color='YlOrRd',
       fill_opacity='0.7',
       line_opacity='0.2',
       legend_name='Urgences Medical Potentially Life-Threatening à San Francisco, par quartier')

# display the map
sf_map


df4_fire=df4.loc[df4['Call Category']=='Incidents']
sf = df4_fire.groupby('Neighborhooods - Analysis Boundaries').count()
sf = pd.DataFrame(sf,columns=['Call Category'])  # remove unneeded columns
sf.reset_index(inplace=True)   # default index, otherwise groupby column becomes index
sf.rename(columns={'Neighborhooods - Analysis Boundaries':'Neighborhood','Call Category':'Count'}, inplace=True)
sf.sort_values(by='Count', inplace=True, ascending=False)


# San Francisco latitude and longitude values
latitude = 37.77
longitude = -122.42
sf_neighborhood_geo = 'Analysis Neighborhoods.geojson'

# Create map
sf_map = folium.Map(
       location=[latitude,longitude],
       zoom_start=12)

 
# Use json file  TEST based on class
sf_map.choropleth(
       geo_data=sf_neighborhood_geo,
       data=sf,
       columns=['Neighborhood','Count'],
       key_on='feature.properties.nhood',
       fill_color='YlOrRd',
       fill_opacity='0.7',
       line_opacity='0.2',
       legend_name='Urgences Incidents à San Francisco, par quartier')

# display the map
sf_map


df4_fire=df4.loc[df4['Call Category']=='Alarms']
sf = df4_fire.groupby('Neighborhooods - Analysis Boundaries').count()
sf = pd.DataFrame(sf,columns=['Call Category'])  # remove unneeded columns
sf.reset_index(inplace=True)   # default index, otherwise groupby column becomes index
sf.rename(columns={'Neighborhooods - Analysis Boundaries':'Neighborhood','Call Category':'Count'}, inplace=True)
sf.sort_values(by='Count', inplace=True, ascending=False)


# San Francisco latitude and longitude values
latitude = 37.77
longitude = -122.42
sf_neighborhood_geo = 'Analysis Neighborhoods.geojson'

# Create map
sf_map = folium.Map(
       location=[latitude,longitude],
       zoom_start=12)

 
# Use json file  TEST based on class
sf_map.choropleth(
       geo_data=sf_neighborhood_geo,
       data=sf,
       columns=['Neighborhood','Count'],
       key_on='feature.properties.nhood',
       fill_color='YlOrRd',
       fill_opacity='0.7',
       line_opacity='0.2',
       legend_name='Urgences Alarms à San Francisco, par quartier')

# display the map
sf_map

	Call Date	Received DtTm	Address	Zipcode of Incident	Battalion	Station Area	Box	Final Priority	Neighborhooods - Analysis Boundaries	Location	shape	Neighborhoods - Analysis Boundaries	year	month	day	minute	weekday	Call Category
2949367	2012-01-01	2012-01-01 00:00:57	0 Block of WASHINGTON ST	94111.0	B01	13	1116	2	Financial District/South Beach	(37.796450111202, -122.396164646182)	POINT (-122.396164646182 37.796450111202)	8.0	2012	1	1	0	Sunday	Medical Non Life-Threatening
2920610	2012-01-01	2012-01-01 00:02:05	GEARY BL/CLEARY CT	94109.0	B04	3	3323	2	Western Addition	(37.785061737122, -122.42643012773)	POINT (-122.42643012773 37.785061737122)	39.0	2012	1	1	2	Sunday	Medical Non Life-Threatening
3003150	2012-01-01	2012-01-01 00:02:27	HAIGHT ST/LAGUNA ST	94102.0	B02	36	3415	3	Hayes Valley	(37.772734311086, -122.425512056036)	POINT (-122.425512056036 37.772734311086)	9.0	2012	1	1	2	Sunday	Fire
2990241	2012-01-01	2012-01-01 00:02:53	WALLER ST/LAGUNA ST	94102.0	B02	36	3415	3	Hayes Valley	(37.771795346154, -122.425322486842)	POINT (-122.425322486842 37.771795346154)	9.0	2012	1	1	2	Sunday	Fire
3002275	2012-01-01	2012-01-01 00:03:55	1100 Block of FOLSOM ST	94103.0	B02	29	2322	3	South of Market	(37.775260329348, -122.409820946894)	POINT (-122.409820946894 37.775260329348)	34.0	2012	1	1	3	Sunday	Medical Potentially Life-Threatening

	Call Date	Received DtTm	Address	Zipcode of Incident	Battalion	Station Area	Box	Final Priority	Neighborhooods - Analysis Boundaries	Location	shape	Neighborhoods - Analysis Boundaries	year	month	day	minute	weekday	Call Category
2949367	2012-01-01	2012-01-01 00:00:57	0 Block of WASHINGTON ST	94111.0	B01	13	1116	2	Financial District/South Beach	(37.796450111202, -122.396164646182)	POINT (-122.396164646182 37.796450111202)	8.0	2012	1	1	0	Sunday	Medical Non Life-Threatening
2920610	2012-01-01	2012-01-01 00:02:05	GEARY BL/CLEARY CT	94109.0	B04	3	3323	2	Western Addition	(37.785061737122, -122.42643012773)	POINT (-122.42643012773 37.785061737122)	39.0	2012	1	1	2	Sunday	Medical Non Life-Threatening
2933160	2012-01-01	2012-01-01 00:10:24	3RD ST/MARKET ST	94103.0	B01	1	1241	2	Financial District/South Beach	(37.787641847607, -122.403430803415)	POINT (-122.403430803415 37.787641847607)	8.0	2012	1	1	10	Sunday	Medical Non Life-Threatening
2990728	2012-01-01	2012-01-01 00:17:16	1100 Block of FILLMORE ST	94115.0	B05	5	3535	2	Western Addition	(37.780021494942, -122.432114423941)	POINT (-122.432114423941 37.780021494942)	39.0	2012	1	1	17	Sunday	Medical Non Life-Threatening
2967994	2012-01-01	2012-01-01 00:18:23	500 Block of BARNEVELD AVE	94124.0	B10	9	6427	2	Bayview Hunters Point	(37.74044593571, -122.404296869791)	POINT (-122.404296869791 37.74044593571)	1.0	2012	1	1	18	Sunday	Medical Non Life-Threatening

	month
8	33133
3	32988
1	32961
5	32833
7	32545
6	31882
4	31711
2	30213
12	30181
10	29914
9	29444
11	28126

	Effectif	Mois
8	33133	8
3	32988	3
1	32961	1
5	32833	5
7	32545	7
6	31882	6
4	31711	4
2	30213	2
12	30181	12
10	29914	10
9	29444	9
11	28126	11

Projet Machine Learning avec Python :

Prédiction des incidents à San Francisco

Partie 1 : Opérations préliminaires sur la base et visualisation des données

1. Opérations préliminaires sur la base de données¶

2.1. Visualisation des données temporelles¶

2.2. Visualisation des données géographiques¶